Beyond Life's Essential 8: optimizing cardiovascular health metrics to predict mortality

Background: Cardiovascular health (CVH), as assessed by the American Heart Association's Life's Essential 8 (LE8), is strongly associated with mortality risk. However, whether rescoring or weighting individual CVH components improves mortality prediction remains unclear. Methods: Using data from the 2005-2018 National Health and Nutrition Examination Survey, we examined the associations between CVH categories and risks of all-cause and cardiovascular disease (CVD) mortality. We compared three CVH scoring approaches: the original LE8 model, a rescored model with recalibrated eight metrics, and a weighted model assigning relative importance to each metric. Cox proportional hazards models adjusted for confounders estimated hazard ratios. Model performance was evaluated by C-statistic and net reclassification improvement. Results: Among 32,076 US adults followed for a median of 7.5 years, higher CVH was consistently associated with lower all-cause and CVD mortality risks across all models. Compared to individuals with low CVH, individuals with high CVH had 58 %-78 % lower all-cause mortality risk and 64 %-87 % lower CVD mortality. For CVD mortality, the rescored model improved risk reclassification, while the weighted model improved discrimination. Compared to the original model, both rescored and weighted models are with modest improvements in all-cause mortality prediction. Weighting revealed substantial variation in the contribution of individual CVH components to mortality risk. Conclusions: Higher CVH is strongly protective against mortality. Refining LE8 scoring through rescoring and weighting can enhance mortality risk discrimination and reclassification, supporting improved CVH assessment for targeted prevention.